On Oct 3, 2013, at 4:31 PM, Wes Hardaker <[email protected]> wrote:

> Warren Kumari <[email protected]> writes:
> 
> Looks like a good start and thanks for writing it Olafur!  But it's not
> ready for publication without some cleanup.  Too many standard-things
> are missing, such as acronym expansion and I'd feel guilty passing it
> to the RFCEditors without us having caught such things.
> 
> Nits:
> 
> 1)
>   handle any case use in the input> The expectation is that by using
>                              ^^^^^^
>                              ??????
> 
> 2) my alternate suggested should descriptions (I like the acronyms 
> themselves):
> 
>    +-------+----------+-------------------------------------+-----------+
>    | Value | Acronym  | Short Description                   | Reference |
>    +-------+----------+-------------------------------------+-----------+
>    |     0 | PKIX-TA  | PKIX Validated CA Reference         | [RFC6698] |
>    |     1 | PKIX-EE  | PKIX Validated End-Entity Reference | [RFC6698] |
>    |     2 | DANE-TA  | Validation TA Reference             | [RFC6698] |
>    |     3 | DANE-EE  | Domain-issued certificate Reference | [RFC6698] |
>    | 4-254 |          | Unassigned                          |           |
>    |   255 | PrivCert | Reserved for Private Use            | [RFC6698] |
>    +-------+----------+-------------------------------------+-------------+
> 
>                     Table 1: TLSA Certificate Usages
> 
> 3) intro text (1 sentence?) to sections needed and expansion of other
>   acronyms earlier in the document (PKIX, EE, )
> 
> 4) ideally space align the section 3 records
> 
>                             inserted one space here
>                             v
>   _666._tcp.first.example.   TLSA PKIX-CA CERT SHA2-512 {blob}
>   _666._tcp.second.example.  TLSA DANE-TA SPKI SHA2-256 {blob}
> 
> 5) security considerations
> 
>   There is definitely something to consider if someone publishes both
>   name records along with number records, and the client only parses
>   number records.  What happens with this:
> 
>   _666._tcp.first.example.   TLSA 3       1    1        {blob}
>   _666._tcp.first.example.   TLSA DANE-TA SPKI SHA2-256 {blob}
> 
>   Something needs to be said for that case; what would an existing
>   implementation do?  drop both? take one?  Either way, it should be
>   discussed/mentioned.
> 

<no hats>
Um, I'm a little confused (or you are, but probably it is me…)

I'm not quite sure what you are menacing by client here.  
If it is something like a stub resolver / application, it doesn't really see 
either of the above -- it just gets a big bag-o-bits.

If you are talking about someone putting this in a zone file,  whatever eats 
the zone file would continue to do what it does with any other record.
Example:
example            TYPE16 "foo"
example            TXT    "bar"

These are two text records, one with content "foo" and one with content "bar"
If someone does:
_666._tcp.first.example.   TLSA 3       1    1        {blob}
_666._tcp.first.example.   TLSA DANE-TA SPKI SHA2-256 {blob}

it is the same as if they had entered:
_666._tcp.first.example.   TLSA 3   1   1        {blob}
_666._tcp.first.example.   TLSA 2  1 1  {blob}
(or, if they have the same usage / selector / selector / match and {blob} then 
it is the same as if they had just entered the info twice).

If whatever eats the zone file is too old to understand the new enums / names, 
then it will continue to do what it does now with unknown garbage:

_666._tcp   TLSA COOKIEMONSTER BARNIE FRED   1234…

wkumari@vimes:~/tmp$ named-checkzone example.com example.com
dns_rdata_fromtext: example.com:26: near 'COOKIEMONSTER': not a valid number
example.com: file does not end with newline
zone example.com/IN: loading from master file example.com failed: not a valid 
number
zone example.com/IN: not loaded due to errors.

I suspect that I'm missing something / we are talking past each other…

W


> -- 
> Wes Hardaker
> Parsons
> 

--
"Let's just say that if complete and utter chaos was lightning, he'd be the 
sort to stand on a hilltop in a thunderstorm wearing wet copper armour and 
shouting 'All gods are bastards'."

    -- Rincewind discussing Twoflower (Terry Pratchett, The Colour of Magic)


_______________________________________________
dane mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dane

Reply via email to